... | ... | @@ -21,7 +21,9 @@ and less account for half the time steps. |
|
|
Richard and I initially speculated that this might best be explained by the fact that the number
|
|
|
of tasks, a proxy for cells, is roughly the same for these steps (seemed that way
|
|
|
to the eye). So we could be processing the same number of cells for different numbers
|
|
|
of particles... Turns out to be not probably the case as a plot of active tasks per step
|
|
|
against updated particles seems linear:
|
|
|
of particles... So a plot of active tasks per step against updated particles:
|
|
|
|
|
|
![eagle_25-12cores-fig3](/uploads/34c0c78e99bc68bb6bd45f7ee46b1d75/eagle_25-12cores-fig3.png) |
|
|
\ No newline at end of file |
|
|
![eagle_25-12cores-fig3](/uploads/34c0c78e99bc68bb6bd45f7ee46b1d75/eagle_25-12cores-fig3.png)
|
|
|
|
|
|
Looks quite linear down to ~500 particles, when we loose scaling, but that is still 40%
|
|
|
of all steps. |
|
|
\ No newline at end of file |